The time required to access a page on a web site typically varies across different devices, browsers, and networks. Web page access time may be used to evaluate the operation of these devices, browsers, and networks. For example, a web page download time Key Performance Indicator (KPI) may be defined as the time elapsed from the moment a user enters a web page URL into a browser until the time when all artifacts related to that web page have downloaded and page rendering by the browser is complete.
The web page download time KPI may be of interest to mobile network operators for several reasons. The web page download time KPI is an objective measurement of a mobile subscriber's experience and satisfaction, particularly because web browsing is the number one service used in mobile networks. The web page download time KPI allows network operators to detect problems with particular handset types or handset software versions in the network and to request handset manufacturers to provide fixes for those problems to reduce customer care calls. Mobile network operators may also use the web page download time KPI to examine and fine-tune hosted websites and portals for optimal user experience. For example, thresholds and alarms can be set to detect when the average performance for hosted or third-party web pages is degraded, which may indicate network-related issues.
The web page download time KPI can be used to estimate uplink and downlink throughput rates that are being provided to subscribers for web browsing services. Additionally, mobile network operators may use web page download time KPIs to pinpoint network locations where service is degraded.
Currently, the most direct and precise method to measure web page download time KPIs is embedding a software agent in a web browser. The software agent then reports web page download times to a centralized data collection entity. Because the browser knows exactly when the web page download started and ended and knows how many objects were retrieved, a browser agent can provide highly accurate web page download time KPIs. However, this approach is not practical because it requires software agents adapted for a multitude of handsets, browsers, and user agents. Additionally, user privacy concerns make it an unrealistic approach for network-wide use.
Embedded page scripting, such as JavaScript, may be used to embed logic in a web page to trigger sending web page download statistics to the server. Once the web page completes downloading and rendering, the browser automatically executes commands, such as calculating and sending the web page download statistics. However, this approach cannot be used for web pages that lack these embedded scripts or on handsets or laptops that either lack JavaScript capability. Additionally, these statistics cannot include throughput and other measurements because TCP stack control frames and retransmissions are not visible to the browser.
These and other known approaches for measuring web page download time are not practicable in a passive monitoring system that does not have access to subscriber browser data. Accordingly, other methods of estimating web page download time must be used by network operators to determine the web page download time KPIs for different subscribers.